Better students lead to higher teacher ratings
A new study examines classroom observations as a method of teacher evaluation and finds that teachers with high achieving students get better ratings. In other words, a principal, department chair, or some other reviewer is likely to report that an AP Calculus teacher performs better than her colleague who teaches remedial math.
Last month, in my discussion of the AERA statement on value-added models of teacher evaluation (VAM), I argued that we can only evaluate VAM fairly if we consider its merits and shortcomings vis a vis other widely used methods like classroom observation. Steinberg and Garrett's study shows why. Critics have charged that teachers who teach better students will get better VAM scores, and teachers who do not will be unfairly judged inferior. To be blunt, I strongly suspect such criticisms are misinformed at best. VAM's biggest strength is that it controls for previous student achievement so that teachers are only measured on the progress their students make. If anything, VAM should reward those who teach previously low achieving students, because these students have more room to grow.
What Steinberg and Garrett have shown is that the charge that higher achieving students artificially inflates teacher ratings is better leveled at classroom observation. High achieving students are more likely to appear engaged, immediately understand new concepts, and do all of the other things that qualitative evaluation rubrics reward. This study finds that observers are likely to incorrectly attribute student characteristics to teacher performance, which is why we need VAM.